With the tremendous growth in the use of biomedical literature, practitioners have to analyze day to day generated 3D medical images of sonography, MRI scan, CT scan. The idea behind this survey is to study different 3D image concepts and processing methods which will help to create model of 3D image analysis, retrieval in the field of medical domain. A new technique like Hadoop'sMapReduce is used to achieve parallelism in Image processing methods

ificial intelligence has recently made a disruptive impact in medical imaging by successfully automatizing expert-level diagnostic tasks. However, replicating human-made decisions may inherently be biased by the fallible and dogmatic nature of human experts, in addition to requiring prohibitive amounts of training data. In this paper, we introduce an unsupervised deep learning architecture particularly designed for OCT representations for unbiased, purely data-driven biomarker discovery. We developed artificial intelligence technology that provides biomarker candidates without any restricting input or domain knowledge beyond raw images. Analyzing 54,900 retinal optical coherence tomography (OCT) volume scans of 1094 patients with age-related macular degeneration, we generated a vocabulary of 20 local and global markers capturing characteristic retinal patterns. The resulting markers were validated by linking them with clinical outcomes (visual acuity, lesion activity and retinal morphology) using correlation and machine learning regression. The newly identified features correlated well with specific biomarkers traditionally used in clinical practice (r up to 0.73), and outperformed them in correlating with visual acuity ( R^2 = 0.46 compared to R^2 = 0.29 for conventional markers), despite representing an enormous compression of OCT imaging data (67 million voxels to 20 features). In addition, our method also discovered hitherto unknown, clinically relevant biomarker candidates. The presented deep learning approach identified known as well as novel medical imaging biomarkers without any prior domain knowledge. Similar approaches may be worthwhile across other medical imaging fields.

The identification and quantification of markers in medical images is critical for diagnosis, prognosis, and disease management. Supervised machine learning enables the detection and exploitation of findings that are known a priori after annotation of training examples by experts. However, supervision does not scale well, due to the amount of necessary training examples, and the limitation of the marker vocabulary to known entities. In this proof-of-concept study, we propose unsupervised identification of anomalies as candidates for markers in retinal optical coherencetomography (OCT) imaging data without a constraint to a priori definitions. We identify and categorize marker candidates occurring frequently in the data and demonstrate that these markers show a predictive value in the task of detecting disease. A careful qualitative analysis of the identified data driven markers reveals how their quantifiable occurrence aligns with our current understanding of disease course, in early-and late age-related macular degeneration (AMD) patients. A multi-scale deep denoising autoencoder is trained on healthy images, and a one-class support vector machine identifies anomalies in new data. Clustering in the anomalies identifies stable categories. Using these markers to classify healthy-, early AMD- and late AMD cases yields an accuracy of 81.40%. In a second binary classification experiment on a publicly available data set (healthy versus intermediate AMD), the model achieves an area under the ROC curve of 0.944.

We propose an unsupervised object category learning approach, where the output representations improve classification performance too. The contribution is threefold: we integrate a ’Network in Network’ (NIN) approach in unsupervised learning, improving the representational power of the network for local patches by replacing linear filters with micro convolutional networks. We learn receptive fields to connect layers of the network sparsely, and we propose a new encoding function that introduces sparsity in a natural way and avoids the necessity of parameter tuning. The learned model generates a feature representation of images, used for unsupervised category learning. Results demonstrate that the obtained image categories reflect true object categories well. In addition, experimental results on classification tasks show superiority of the proposed approach in comparison with an unsupervised stateof-the-art learning architecture.

The identification and quantification of markers in medical images is critical for diagnosis, prognosis and management of patients in clinical practice. Supervised- or weakly supervised training enables the detection of findings that are known a priori. It does not scale well, and a priori definition limits the vocabulary of markers to known entities reducing the accuracy of diagnosis and prognosis. Here, we propose the identification of anomalies in large-scale medical imaging data using healthy examples as a reference. We detect and categorize candidates for anomaly findings untypical for the observed data. A deep convolutional autoencoder is trained on healthy retinal images. The learned model generates a new feature representation, and the distribution of healthy retinal patches is estimated by a one-class support vector machine. Results demonstrate that we can identify pathologic regions in images without using expert annotations. A subsequent clustering categorizes findings into clinically meaningful classes. In addition the learned features outperform standard embedding approaches in a classification task.

The automated annotation of bones that are visible in CT images of the skeleton is a challenging task which has, so far, been approached for only certain subregions of the skeleton, such as the spine or hip. In this paper, we propose a novel annotation algorithm for automatically identifying structures and substructures in the whole skeleton. Our annotation algorithm makes use of recent advances in anatomical landmarks detection and is capable of generalising local information about landmarks to a dense label map of the full skeleton by anatomical triangulation. We follow a recognition approach that combines the use of distance-based features for measuring Euclidean and geodesic distances to a few given landmark locations, a parts-based model that is disambiguating anatomical substructures, and an iterative scheme for considering distances to the previously detected structures and, hence, to a dense set of anatomical reference points. We propose an annotation protocol for 136 substructures of the skeleton and test our annotation algorithm on 18 CT images. On average, we obtain a Dice score of 90.54.

Bone age estimation (BAE) is an important procedure in forensic practice which recently has seen a shift in attention from X-ray to MRI based imaging. To automate BAE from MRI, localization of the joints between hand bones is a crucial first step, which is challenging due to anatomical variations, different poses and repeating structures within the hand. We propose a landmark localization algorithm using multiple random regression forests, first analyzing the shape of the hand from information of the whole image, thus implicitly modeling the global landmark configuration, followed by a refinement based on more local information to increase prediction accuracy. We are able to clearly outperform related approaches on our dataset of 60 T1-weighted MR images, achieving a mean landmark localization error of 1.4 ± 1.5mm, while having only 0.25% outliers with an error greater than 10mm.

The enormous amount of visual data in Picture Archival and Communication Systems (PACS) and in the medical literature is growing exponentially. In the proposed demo, the medical image search of the KHRESMOI project is presented to solve some of the challenges of medical data management and retrieval. The system allows searching for visual information by combining content–based image retrieval (CBIR) and text retrieval in several languages using semantic concepts. 3D visual retrieval in internal hospital sources is supported by marking volumes of interest (VOI) in the data and connection to the medical literature are established to allow further investigating interesting cases. The system is demonstrated on 5TB of radiology reports with associated images and articles of the biomedical literature with over 1.7M images.

Prenatal neuroimaging requires reference models that reflect the normal spectrum of fetal brain development, and summarize observations from a representative sample of individuals. Collecting a sufficiently large data set of manually annotated data to construct a comprehensive in vivo atlas of rapidly developing structures is challenging but necessary for large population studies and clinical application. We propose a method for the semi-supervised learning of a spatio-temporal latent atlas of fetal brain development, and corresponding segmentations of emerging cerebral structures, such as the ventricles or cortex. The atlas is based on the annotation of a few examples, and a large number of imaging data without annotation. It models the morphological and developmental variability across the population. Furthermore, it serves as basis for the estimation of a structures' morphological age, and its deviation from the nominal gestational age during the assessment of pathologies. Experimental results covering the gestational period of 20-30 gestational weeks demonstrate segmentation accuracy achievable with minimal annotation, and precision of morphological age estimation. Age estimation results on fetuses suffering from lissencephaly demonstrate that they detect significant differences in the age offset compared to a control group. (C) 2013 Published by Elsevier B.V.

Fully automatic localization of anatomical structures in 2D and 3D radiological data sets is important in both computer aided diagnosis, and the rapid automatic processing of large amounts of data. We present a simple, accurate and fast approach with low computational complexity to find anatomical landmarks, based on a multi-scale regression codebook of informative image patches and encoded landmark contexts.From a set of annotated training volumes the method captures the appearance of landmarks over several scales together with relative positions of neighboring landmarks and a spatial distribution model. During multi-scale search in a target volume, starting from the coarsest level, each landmark model predicts all landmark positions it has encoded, with the median of all predictions yielding the final prediction for each scale.We present results on two challenging data sets (hand radiographs and hand CTs), where our method achieves comparable accuracy to the state of the art with substantially improved run-time.

Atlases have a tremendous impact on the study of anatomy and function, such as in neuroimaging, or cardiac analysis. They provide a means to compare corresponding measurements across populations, or model the variability in a population. Current approaches to construct atlases rely on examples that show the same anatomical structure (e. g., the brain). If we study large heterogeneous clinical populations to capture subtle characteristics of diseases, we cannot assume consistent image acquisition any more. Instead we have to build atlases from imaging data that show only parts of the overall anatomical structure. In this paper we propose a method for the automatic contruction of an un-biased whole body atlas from so-called fragments. Experimental results indicate that the fragment based atlas improves the representation accuracy of the atlas over an initial whole body template initialization.

We propose an approach which allows to localize anatomical landmarks in radiological datasets given only a single manual annotation and set of un-annotated example images. Using top-down image patch regression to obtain potential landmark candidates in the set of training images, a model of the anatomical structure is incrementally enlarged, starting from the single, annotated image, until it encompasses the entire training set. The obtained model then allows to perform highly accurate anatomical structure localization on test data. We report preliminary results on a set of 2D radio-graphs, with a median/mean localization residual of 0.92 mm/1.30 mm. The approach yields very promising localization results, suggesting that is possible to eliminate the tedious manual annotation process still required by state of the art localization approaches.

There is increasing interest in and need for innovative solutions to medical search. In this paper we present the EU funded Khresmoi medical search and access system, currently in year 3 of 4 of development across 12 partners . The Khresmoi system uses a component based architecture housed in the cloud to allow for the development of several innovative applications to support target users medical information needs. The Khresmoi search systems based on this architecture have been designed to support the multilingual and multimod al information needs of three target groups the general public, general practitioners and consultant radiologists. In this paper we focus on the presentation of the systems to support the latter two groups using semantic, multilingual text and image based (including 2D and 3D radiology images) search.

Abstract. Fetal magnetic resonance imaging (MRI) is a rapidly growing field of research for its potential to study brain development in utero. However, in contrast to adult studies automatic brain extraction and orientation is not yet solved, but remains challenging in wide field of view raw fetal MRI volumes. This has limited research to small scale studies. This paper presents an automatic fetal brain extraction and orientation framework to remove this limitation. The method consists of a two-phase random forest classifier, and an approximate high-order Markov random field solution, that results in a brain mask for an MRI stack. The resulting extraction achieves 98% detection rate with 88% mean sensitivity when validated on a set of cases aged between 18-30.2 gestational weeks (GW), supporting a robust pipeline to automated fetal MRI processing techniques.

A prototype of a content based information retrieval system for clinical images is presented. It is targeted at the user group of radiologists working on diagnosing new cases amongst other scenarios. While still being in an early state, the system features sophisticated image retrieval mechanisms and a comprehensive and versatile user interface. CBIR is performed using several types of visual features aggregated over supervoxels and state-of-the art indexing regimes. The user interface uses an agent based framework infrastructure which is easily extensible and suited for complex tasks with difficult information needs [Beckers et al., 2012b]. The system is one main part of the EU-project Khresmoi finishing in August 2014.

Khresmoi is a European Integrated Project developing a multilingual multimodal search and access system for medical and health information and documents. It addresses the challenges of searching through huge amounts of medical data, including general medical information available on the internet, as well as radiology data in hospital archives. It is developing novel semantic search and visual search techniques for the medical domain. At the MIE Village of the Future, Khresmoi proposes to have two interactive demonstrations of the system under development, as well as an overview oral presentation and potentially some poster presentations.

The accurate localization of anatomical landmarks is a challenging task, often solved by domain specific approaches. We propose a method for the automatic localization of landmarks in complex, repetitive anatomical structures. The key idea is to combine three steps: (1) a classifier for pre-filtering anatomical landmark positions that (2) are refined through a Hough regression model, together with (3) a parts-based model of the global landmark topology to select the final landmark positions. During training landmarks are annotated in a set of example volumes. A classifier learns local landmark appearance, and Hough regressors are trained to aggregate neighborhood information to a precise landmark coordinate position. A non-parametric geometric model encodes the spatial relationships between the landmarks and derives a topology which connects mutually predictive landmarks. During the global search we classify all voxels in the query volume, and perform regression-based agglomeration of landmark probabilities to highly accurate and specific candidate points at potential landmark locations. We encode the candidates' weights together with the conformity of the connecting edges to the learnt geometric model in a Markov Random Field (MRF). By solving the corresponding discrete optimization problem, the most probable location for each model landmark is found in the query volume. We show that this approach is able to consistently localize the model landmarks despite the complex and repetitive character of the anatomical structures on three challenging data sets (hand radiographs, hand CTs, and whole body CTs), with a median localization error of 0.80 mm, 1.19 mm and 2.71 mm, respectively.